© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 1978
    Prev
    Next
    INDEX
    Explanations

    worth compared to costs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    alen
    -0.24
     Mud
    -0.24
    åħĪéĶĭ
    -0.24
    离
    -0.23
     visionary
    -0.23
    visibility
    -0.23
     sprung
    -0.23
     strt
    -0.22
    èĬ¸
    -0.21
    款
    -0.21
    POSITIVE LOGITS
    è¿ĺä¸įå¦Ĥ
    0.50
    åĪĴç®Ĺ
    0.46
    æĢ§ä»·æ¯Ķ
    0.42
     outweigh
    0.38
    çļĦæĪIJæľ¬
    0.33
     compared
    0.32
    _costs
    0.31
     вÑĭгод
    0.31
     costs
    0.31
    ä¸įå¦Ĥ
    0.30
    Activations Density 0.092%

    No Known Activations